The invention relates to a method for the recording of data onto a moving recording medium and for the reproduction of the data free of loss from the recording medium whereby on the recording side a data compression with differing degrees of compression and on the reproduction side a data expansion is carried out.
Data compression and data expansion are frequently used in the transmission of information in order to, for example, increase the capacity of transmission channels or the storage capacity of recording media. For example, on the so-called Mini-Disk, a further development of the compact disk, the data are recorded in compressed form.
In data compression the data to be stored or to be transmitted are compressed using a suitable algorithm. This means, in other words, that the amount of data to be transmitted or stored respectively is reduced in order to increase the transmission capacity of the transmission channel or, respectively, the storage capacity of the recording medium.
On the reproduction or receiving side respectively an expander generates the original data from the reduced amount of data using the inverse algorithm.
The aim of data compression is, as mentioned above, to reduce the amount of data as much as possible without creating loss of data in order to better utilize the transmission capacity of a transmission channel or the storage capacity of a recording medium.
It is known, with the recording of data, to read the data to be recorded into an intermediate memory prior to the recording. Likewise, it is known, with the reproducing of data, to first read the data to be reproduced into an intermediate memory prior to the reproduction. However, if the intermediate memory overflows loss of data occurs not only with the recording but also with the data reproduction. In order to keep such loss of data as small as possible, upon recording the data, the most important bits are first generated by means of the algorithm and stored so that less important bits which do not materially contribute to the regeneration of the original data signal are stored last or, respectively, get lost upon overflow of the intermediate memory. In the latter case a recording of the data free of loss is no longer possible.
Similarly, upon reading out the data the most important bits are read out firstly and lastly those bits which do not materially contribute to the regeneration of the original data signal. Here too, loss of data occurs upon overflow of the intermediate memory as the less important bits get lost through the overflow.
To avoid loss of data through overflow of the intermediate memory intermediate memories with sufficiently large storage capacity are required.